diff options
Diffstat (limited to 'README.TXT')
| -rw-r--r-- | README.TXT | 27 |
1 files changed, 17 insertions, 10 deletions
@@ -1,4 +1,6 @@ -* Contact information. +-*-org-*- + +* Contact information. Any feedback will be appreciated. You can email us at Daniel M. German <dmg@uvic.ca> and Yuki Manabe <y-manabe@ist.osaka-u.ac.jp> @@ -25,11 +27,16 @@ http://turingmachine.org/~dmg/papers/dmg2010ninka.pdf If you use Ninka for research purposes, we would appreciate you cite the above paper. +* Contributors + +- Anthony Kohan for writing the excel and sqlite backends. +- Armijn Hemel from Tjaldur Software Governance Solutions for multiple bug reports and suggestions + * License - + Except for the directories comments and splitter, Ninka is licensed under the GPLv2+ - + Copyright (C) 2009-2010 Yuki Manabe and Daniel M. German This program is free software: you can redistribute it and/or modify @@ -69,7 +76,7 @@ Perl version 5 Ninka uses a pipe model (see below). Each step of the "pipe" creates a file, but -ninka.pl [options] [filename] +ninka.pl [options] [filename] Available options -v verbose @@ -107,26 +114,26 @@ It will create five files: - Licenses - Unmatched sentences in *.senttok that were not matched - + * Ninka model Ninka uses a pipe-model. Each stage of the pipe does something very specific: - 1. Comment extractor. + 1. Comment extractor. - directory: extComments - command: extComments.pl, might use comments (included in distribution) - + - Purpose: Extracts top comments of source code. If no comment extractor is known for the language, then extracts top lines from source (currently 700) - Creates <filename>.comments file 2. Split sentences in comments - + - directory: splitter - command: splitter.pl @@ -151,7 +158,7 @@ Ninka uses a pipe-model. Each stage of the pipe does something very specific: 4. Tokenizes sentences - Directory senttok - + - command: senttok.pl - Purpose: It creates a file that corresponds to the recognized @@ -168,7 +175,7 @@ Ninka uses a pipe-model. Each stage of the pipe does something very specific: - Purpose: looks at the sequence of sentence tokens and outputs the licenses found - Output: <filename>.license - + The script ninka.pl takes care of all these steps, and optionally removes intermediary files, and writes to the stdout the licenses found. |